Diabetes increases the risk for stroke and worsens outcomes including physical disability and post-stroke cognitive impairment (PSCI). Diabetic patients are more likely to bleed into the brain (hemorrhagic transformation, HT), an important complication of ischemic stroke, that is poorly studied in females. The lack of understanding on how increased bleeding occurs and influences the restorative and regenerative processes within the neurovascular networks hindered the development of new therapeutic strategies for stroke recovery. While clinically it is known that women suffer more from poor outcomes and PSCI, necessitating long-term nursing care, the inadequate inclusion of female animals coupled with the limited use of diabetic models in preclinical stroke research has further deepened this gap. Our objective is to address this vast knowledge gap by focusing on the mechanisms and consequences of increased HT in diabetic females. We made the novel observations that: 1) young diabetic female rats lose the neuroprotection typically seen in control female animals and develop greater HT than in controls and even diabetic male rats; 2) matrix metalloprotease (MMP)-3, an enzyme known to cause HT and to be regulated by toll like receptor 4 (TLR4), is increased to a greater degree in cerebral microvessels of female diabetic rats; 3) while male diabetic animals show significant loss of cerebrovasculature by activation of multiple cell death pathways in the recovery period, female diabetic animals do not, but rather undergo phenotypic changes in endothelial cells resembling endothelial- mesenchymal transition, EndMT, a process associated with scarring and impaired healing, and 4) in the long- term, diabetes worsens sensorimotor and cognitive recovery in both sexes. Thus, we hypothesize that endothelial (e)TLR4 has a dual role in amplified vascular injury and compromised vascular restoration & recovery in females with diabetes: eTLR4-mediated increase in MMP3 activity amplifies HT which in turn sustains TLR4 activation leading to transforming growth factor (TGF)-?-facilitated EndMT ? loss of neurovascular unit (NVU) integrity ? poor functional recovery. 3 aims will test the subhypotheses that: 1) eTLR4-mediated MMP3 activation amplifies HT and worsens stroke outcomes in diabetes; 2) sustained eTLR4 activation due to HT mediates EndMT resulting in loss of NVU integrity and poor recovery in diabetes; and 3) amplified TGF-? signaling is the underlying mechanism of eTLR4-driven EndMT in diabetes. In 9 translational and mechanistic studies over 5 years, and utilizing rigorous behavioral assessment of control and diabetic female animals, pharmacologic and genetic manipulations in vivo and in vitro, we will advance our understanding of stroke recovery in females, an understudied population in ischemic stroke, and identify EndMT prevention as a new promising therapeutic tactic for not only stroke but also for vascular contributions to cognitive impairment and dementia (VCID) spectrum of diseases which include PSCI.